Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PaddlePaddle
Paddle
提交
eb7d8754
P
Paddle
项目概览
PaddlePaddle
/
Paddle
大约 1 年 前同步成功
通知
2298
Star
20931
Fork
5422
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1423
列表
看板
标记
里程碑
合并请求
543
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1,423
Issue
1,423
列表
看板
标记
里程碑
合并请求
543
合并请求
543
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
eb7d8754
编写于
5月 18, 2018
作者:
Q
Qiao Longfei
提交者:
daminglu
5月 17, 2018
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
add trainer.stop and fix a bug for train_by_parallel_executor (#10762)
上级
54ae8e45
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
22 addition
and
17 deletion
+22
-17
python/paddle/fluid/tests/book/high-level-api/fit_a_line/test_fit_a_line.py
...d/tests/book/high-level-api/fit_a_line/test_fit_a_line.py
+12
-14
python/paddle/fluid/trainer.py
python/paddle/fluid/trainer.py
+10
-3
未找到文件。
python/paddle/fluid/tests/book/high-level-api/fit_a_line/test_fit_a_line.py
浏览文件 @
eb7d8754
...
...
@@ -57,22 +57,20 @@ def train(use_cuda, train_program, save_dirname):
optimizer
=
fluid
.
optimizer
.
SGD
(
learning_rate
=
0.001
))
def
event_handler
(
event
):
if
isinstance
(
event
,
fluid
.
EndEpochEvent
):
test_metrics
=
trainer
.
test
(
reader
=
test_reader
,
feed_order
=
[
'x'
,
'y'
])
print
test_metrics
'''
...
['25.768919467926025']
['15.343549569447836']
...
'''
if
float
(
test_metrics
[
0
])
<
20.0
:
if
isinstance
(
event
,
fluid
.
EndStepEvent
):
if
event
.
step
==
10
:
test_metrics
=
trainer
.
test
(
reader
=
test_reader
,
feed_order
=
[
'x'
,
'y'
])
print
test_metrics
'''
...
['25.768919467926025']
['15.343549569447836']
...
'''
if
save_dirname
is
not
None
:
trainer
.
save_params
(
save_dirname
)
return
trainer
.
stop
()
trainer
.
train
(
reader
=
train_reader
,
...
...
python/paddle/fluid/trainer.py
浏览文件 @
eb7d8754
...
...
@@ -100,6 +100,7 @@ class Trainer(object):
param_path
=
None
,
place
=
None
,
parallel
=
False
):
self
.
__stop
=
False
self
.
parallel
=
parallel
# 1. we need to generate a framework.Program by calling
# program_func. Reference: fluid.program_guard in
...
...
@@ -210,6 +211,12 @@ class Trainer(object):
'TRAINING_ROLE environment variable must be either TRAINER or PSERVER'
)
def
stop
(
self
):
"""
stop training
"""
self
.
__stop
=
True
def
train
(
self
,
num_epochs
,
event_handler
,
reader
=
None
,
feed_order
=
None
):
"""
Train the model.
...
...
@@ -289,6 +296,8 @@ class Trainer(object):
for
epoch_id
in
range
(
num_epochs
):
event_handler
(
BeginEpochEvent
(
epoch_id
))
for
step_id
,
data
in
enumerate
(
reader
()):
if
self
.
__stop
:
return
begin_event
=
BeginStepEvent
(
epoch_id
,
step_id
)
event_handler
(
begin_event
)
if
begin_event
.
fetch_metrics
:
...
...
@@ -327,9 +336,7 @@ class Trainer(object):
feeder
=
data_feeder
.
DataFeeder
(
feed_list
=
feed_var_list
,
place
=
self
.
place
)
reader
=
feeder
.
decorate_reader
(
reader
,
multi_devices
=
True
)
for
epoch_id
in
range
(
num_epochs
):
self
.
_train_by_any_executor
(
event_handler
,
pe
,
num_epochs
,
reader
)
self
.
_train_by_any_executor
(
event_handler
,
pe
,
num_epochs
,
reader
)
def
_get_parallel_executor
(
self
):
return
getattr
(
self
,
'parallel_executor'
,
None
)
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录